Showing results for 
Search instead for 
Did you mean: 


Multilayer Switching (MLS) offloads a significant portion of routing (packet rewrite) to the hardware. Therefore, it has also been termed switching. The two of the three major components of MLS are the MLS Route Processor (MLS-RP) running Cisco IOS  Software and the MLS-Switching Engine (MLS-SE) running Catalyst OS (CatOS).

With special hardware and software, these components can handle the rewriting of the packet. This frees up vital router resources by allowing other network components to take on some of its functions. MLS is a robust feature, so failures usually arise from misconfiguration, the few software features that are incompatible with MLS or unrelated network performance problems.

There are significant differences between how MLS functions on the Catalyst 6500 and 6000 series and the Catalyst 5500 and 5000 series switches. On the Catalyst 6500 and 6000 series switches, MLS is on by default and used only internally. However, on the Catalyst 5500 and 5000 series switches, MLS is off by default and supports the MLS-RP as the external router.

To determine the reason the MLS software feature is not functioning, perform these steps:

1. Verify that the software and hardware requirements are met by referring to System Requirements to Implement MLS.

    Unlike the Catalyst 5500 and 5000 series, the MLS-RP must be internal for the Catalyst 6500 and 6000 series switches.

2. If a flow has not been created, verify that the Supervisor Engine can see the  MLS-RP (Multilayer Switch Feature Card [MSFC], Route Switch Feature Card [RSFC], Route Switch Module [RSM] or external router) with all expected MAC addresses when you issue the show mls command.

3. Make sure that the MLS-RP is not stuck in ROM Monitor (ROMMON) mode. Check if MLS is enabled on the router by issuing the show mls status command, as shown in this example:

switch2-msfc# show mls status
MLS global configuration status: 
global mls ip: enabled
global mls ipx: enabled
global mls ip multicast: disabled
current ip flowmask for unicast: destination only
current ipx flowmask for unicast: destination only 

4. Verify that the MAC address of the MLS-RP is present when issuing the show cam command.

5. Verify that the source and destination devices on different VLANs are on the same MLS-SE, sharing a common MLS-RP.

6. Verify that the MLS-RP contains an mls rp ip command in its global and interface configuration.

     If it does not, add the command on the MLS-RP.

7. Verify that these features are not enabled on the MLS-RP because they automatically disable MLS:

    If any are present, remove the features.

8. Verify that MLS-SE recognizes the MLS-RP address and the MLS-RP address recognizes the MLS-SE by issuing the show mls rp command, as shown in this example:

switch2-msfc# show mls rp
multilayer switching is globally enabled
mls id is 0010.0b78.4400
!-- This is the MLS ID.
mls ip address
!-- This is the IP address.
mls flow mask is destination-ip
number of domains configured for mls 1

1 management interface(s) currently defined:
vlan 111 on Vlan111
8 mac-vlan(s) configured for multi-layer switching:
mac 0010.0b78.4400
vlan id(s)
111 211 411 412 413 414 415 416 
router currently aware of following 0 switch(es):
!-- MLS-RP does not know of the MLS-SE. No switch IDs currently exist in the domain.

9. Include the MLS-RP address on the MLS-SE.

Configure the MLS-SE switch by issuing the set mls include command, as shown in this example:

console (enable) set mls include

10. Check if the MLS-RP interfaces and the MLS-SE are in the same enabled VLAN Trunk Protocol (VTP) domain. Confirm configuration parameters and restart the MLS process.

11. Verify that the flowmasks match on the MLS-RP and MLS-SE.

12. Check for MLS too many moves error messages continuously seen on the switch.

If there are too many moves error messages, it might be that something is causing the entries to  purge continuously.

Some possible causes of purging the flow table can include these options:

  • Route flapping or any Layer 3 (L3) instability
  • Address Resolution Protocol (ARP) cache changes on the MLS-RP
  • Flowmask changed on the Supervisor Engine
  • Destination VLAN is deleted
  • VLAN interface shutdowns in the MLS-RP

13. To detect and eliminate any physical loop or routing instability on the network, issue the show and spanning-tree commands.

For more information, refer to these documents:

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community:

Quick Links